发明名称 |
PARALLEL RECOVERY BY NON-FAILED NODES |
摘要 |
<p num="1"><br/><br/><br/>Various techniques are described for improving the performance of a shared-<br/>nothing database system in which at least two of the nodes that are running <br/>the shared-nothing database system have shared access to a disk. Specifically, <br/>techniques are provided for recovering the data owned by a failed node using <br/>multiple recovery nodes operating in parallel. The data owned by a failed node <br/>is reassigned to recovery nodes that have access to the shared disk on which <br/>the data resides. The recovery logs of the failed node are read by the <br/>recovery nodes, or by a coordinator process that distributes the recovery <br/>tasks to the recovery nodes.<br/>
|
申请公布号 |
CA2532048(C) |
申请公布日期 |
2013.04.30 |
申请号 |
CA20042532048 |
申请日期 |
2004.07.28 |
申请人 |
ORACLE INTERNATIONAL CORPORATION |
发明人 |
BAMFORD, ROGER;CHANDRASEKARAN, SASHIKANTH;PRUSCINO, ANGELO |
分类号 |
G06F11/20;G06F9/46;G06F9/50;G06F11/00;G06F17/30 |
主分类号 |
G06F11/20 |
代理机构 |
|
代理人 |
|
主权项 |
|
地址 |
|